Machine learning models-based on integration of next-generation sequencing testing and tumor cell sizes improve subtype classification of mature B-cell neoplasms

Abstract

Background: Next-generation sequencing (NGS) panels for mature B-cell neoplasms (MBNs) are widely applied clinically but have yet to be routinely used in a manner that is suitable for subtype differential diagnosis. This study retrospectively investigated newly diagnosed cases of MBNs from our laboratory to investigate mutation landscapes in Chinese patients with MBNs and to combine mutational information and machine learning (ML) into clinical applications for MBNs, especially for subtype classification.

Methods: Samples from the Catalogue Of Somatic Mutations In Cancer (COSMIC) database were collected for ML model construction and cases from our laboratory were used for ML model validation. Five repeats of 10-fold cross-validation Random Forest algorithm was used for ML model construction. Mutation detection was performed by NGS and tumor cell size was confirmed by cell morphology and/or flow cytometry in our laboratory.

Results: Totally 849 newly diagnosed MBN cases from our laboratory were retrospectively identified and included in mutational landscape analyses. Patterns of gene mutations in a variety of MBN subtypes were found, important to investigate tumorigenesis in MBNs. A long list of novel mutations was revealed, valuable to both functional studies and clinical applications. By combining gene mutation information revealed by NGS and ML, we established ML models that provide valuable information for MBN subtype classification. In total, 8895 cases of 8 subtypes of MBNs in the COSMIC database were collected and utilized for ML model construction, and the models were validated on the 849 MBN cases from our laboratory. A series of ML models was constructed in this study, and the most efficient model, with an accuracy of 0.87, was based on integration of NGS testing and tumor cell sizes.

Conclusions: The ML models were of great significance in the differential diagnosis of all cases and different MBN subtypes. Additionally, using NGS results to assist in subtype classification of MBNs by method of ML has positive clinical potential.

Keywords: machine learning (ML); mature B-cell neoplasms (MBNs); next-generation sequencing (NGS); pathological diagnosis; subtype classification.

Figure 1

Grouping procedure and mutation landscape of 849 cases of mature B-cell neoplasms (MBNs). (A) Four groups of 849 MBN cases subcategorized into initial diagnosis and comprehensive diagnosis. (B) The mutation landscape of the top 20 genes detected in 849 MBN cases. (C) Localization and frequencies of 33 KMT2D variants in 26 CLL/SLL cases. *: Stop codon.

Figure 2

Construction and internal validation of machine learning (ML) models based on the COSMIC database. (A) Model feature selection in COSMIC IA. When the model feature number was 30 in COSMIC IA, the model had the highest efficiency (COSMIC IB). (B) Model feature selection in COSMIC IIA. When the model feature number was 16 in COSMIC IIA, the model had the highest efficiency (COSMIC IIB). (C) Model accuracy of COSMIC I (COSMIC IA and IB) and COSMIC II (COSMIC IIA and IIB) in internal validation.

Figure 3

External validation of machine learning (ML) models based on local cohort. (A) Model accuracy of COSMIC II (COSMIC IIA and IIB) in each case group. (B) Model accuracy of COSMIC II (COSMIC IIA and IIB) by subtype in all cases. (C) Model accuracy of COSMIC II (COSMIC IIA and IIB) by subtype in typical cases (Group A). (D) Model accuracy of COSMIC II (COSMIC IIA and IIB) by subtype in refined cases (Group B1). (E) Model accuracy of COSMIC II (COSMIC IIA and IIB) by subtype in further-diagnosed cases (Group B2).

Figure 4

The proportion of cases incorrectly predicted by COSMIC IIB within each subtype of mature B-cell neoplasms (MBNs) based on comprehensive diagnosis. *: Number of incorrectly predicted cases (Incorrectly predicted rate, Total number of predicted cases).